Comparison of machine learning training sampling schemes for induction machine modeling

Abstract

The aim of the paper is to demonstrate the modelingof an induction machine using a chosen machine learningtechnique, followed by a comparison of the training samplingschemes for this technique. A simple 3-phase induction machinewith an axially slitted solid rotor has been selected for the casestudy, where FEM-based program Ansys Electronics Desktophas been used for its calculation. A total of 3 training schemeswere considered and compared with each other for the machinelearning technique. Some of the comparison results are given anddiscussed at the end of this paper. The described methodologycan be used to accelerate the design and optimization of any typeof electrical machine.